complexity of the analytical characterization of polysorbate · 2018-04-02 · pharmaceutical...

Pharmaceutical Development & Manufacturing Sciences

Complexity of the Analytical Characterization of Polysorbate

Case Studies for Degradation Profiling

K. Wuchner, Janssen R&D, Schaffhausen, Switzerland

W. Koch, E. Corradini, A. Hawe, Coriolis Pharma, München, Germany

DISCLAIMER

The perspectives and opinions in this talk are those of the presenter.

CASSS’ AT Europe 2018 in Barcelona, 07 Mar 2018


Functional role of Polysorbates (PS) in biopharmaceutical formulations

• Polysorbate 20 and 80 are the most common non-ionic surfactants in biopharmaceutical

products

• PS have a high surface activity at low concentration and are used as excipient to:

– Prevent protein adsorption

– Protect protein against interfacial stress, surface-induced aggregation and particle

formation

• Concentration range in protein therapeutics from 0.001 to 0.1% (w/v)

– [PS]optimum is experimentally determined by stability studies and stress-studies

– Depends on the stabilization mechanism (competition/ protein-PS interaction)

– Formation of side/degradation products (e.g., PS related particles)

1

➢ Analytical assessment of polysorbate composition, purity and stability are

important during drug product development to ensure sufficient stabilization while

minimizing negative side-effects of PS


Compendial Grade PS

2

➢ EP, USP and JP have harmonized requirements for PS80, which is composed of

multiple fatty acid (FA) esters (PS20: JP is not harmonized)

➢ China enforces compliance to the ChP 2015 and requires “all oleic acid (≥98.0%)”

Grade PS80

Martos et al. (2017), J Pharm Sci, 106:1722-1735

Idealized structure of PS20

Idealized structure of PS80

y+y+z = 20

y+y+z = 20


➢ Optimized UPLC-MS separation of PS20 shows complex chromatogram with

approximately 4000 peaks based on MS spectra (parent (adduct) ions are

counted as one peak)

Complexity of Multi-compendial PSIdealized structure of PS20

3

y+y+z = 20


Complexity of Multi-compendial PS

Sharp peak

4


Complexity of Multi-compendial PS

The sharp peak at 25.7 min is composed by at least ~ 10 peaks based on mass data

Isosorbitane-PEGx10-mono-laurate

with different PEG-ylation levels

➢ High Resolution MS is needed to cope with heterogeneous composition of PS

5


Degradation increases complexity

Potential isomeric forms of sorbitan core

Native Polysorbate

PeroxidesFree FAs, PEGs..

Impurities/byproducts

Free FAs, free PEGs..PeroxidesOxidized FAs, PEGs (intermediates)Short chain organic acids, aldehydes, ketones

Oxidation

Mono-, di, tri-estersFree FAs (FFAs)Free PEGs

Hydrolysis (chemical, enzymatic)

Protein aggregation/precipitationProtein oxidation

Generation of hydrophobic PS degradants (e.g., FFAs)

PS degradation leads toLoss of PS surfactant functionality

6

Various poly(oxyethylene) chains (PEGs) Different fatty acids (FA) esterifiedMono, di-, tri-, tetra-esters

and loss of solubilization capacity

Formation of particles composed of PS degradants


Challenges for analytical characterization of PS

• Polysorbate heterogeneity

complex mixture, markers for degradation pathway are difficult to identify

(needle in haystack)

(reactive) degradants: transient nature and difficult to isolate

no universal PS standard available as reference (e.g. for calibration curves)

• Lack of chromophores

derivatization or universal detectors (CAD, ELSD, MS) required

• Interference of protein and excipients with the analytical methods

sample work-up required (in particular protein removal)

7


Analytical methods for PS in biopharmaceutical formulations

• PS content (validated method with acceptance criteria, R/S)

Fluorescent micelle assay (e.g. NPN assay)

Liquid Chromatography (LC) with charged aerosol detector (CAD) or evaporative light

scattering detector (ELSD)

• PS subclass analysis (characterization method, fast, quantitative)

HPLC-CAD or –ELSD incl. sample preparation for selective isolation of PS

• PS degradants in protein-formulations (characterization method, MS

for identification)

UPLC- MS (positive and/or negative mode) incl. sample preparation for selective

isolation of PS; FA specific methods

• Other methods

Particle characterization methods to determine composition/nature

Peroxide assays, ICP-MS, excipient related methods

8


PS content in biopharmaceutical formulations

Method based on the fluorescent dye N-phenyl-1-

napthylamine (NPN)

• Fluorescence assay can be used to determine PS content

in aqueous solutions

• No sample preparation (just dilution), automated

method (use of well plates)

• Good linearity

NPN

[PS] ≥ CMC: NPN in hydrophobic micellar interior -> high quantum yield

➢ NPN assay is not suitable to quantify PS in each biopharmaceutical

formulation

Caveats of the NPN assay:

NPN interacts with hydrophobic

compounds

interference by certain proteins

and silicone oil

9


PS content and PS class determinations in biopharmaceutical formulations

Fast LC-CAD method to monitor major components of PS

LC-CAD chromatogram for PS20

10

PS can be determined in aqueous solutions (without sample preparation)

Protein needs to be removed e.g., by off-line SPE for biopharmaceutical

formulations

Short cycle times of 15 min

Range: 60 – 1000 µg/mL (~0.006 – 0.1% w/v) PS


Enzymatic degradation studies

LC-CAD method to monitor esterase (0.005 U/mL) induced degradation of 0.04%

PS formulated at pH 6

Mono-ester species degrade faster than Poly-esters

Free PEG fractions increase concomitantly

Part of poly-esters degrades in PS20 but not in PS80 (no change)

No significant difference between Chinese and multicompendial grade PS80

➢ LC-CAD with high sample throughput enabled monitoring of PS

degradation at multiple time points to get insight into kinetics

11


Characterization of PS related molecules and degradants in biopharmaceutical formulations

Isolate PS constituents from Protein/other excipients based on

hydrophobicity

Positive Mode:

Sensitive detection of PS

related molecules

PS degradation pattern of

esters

protein

polysorbateNegative Mode:

Sensitive detection of

FFAs and related

oxidized degradants

12

UPLC-HRMS method

Off-line SPE


Example for multicompendial and Chinese grade PS80 by UPLC-HRMS (pos mode)

13

➢ Differences in overall pattern (base peak chromatogram) is less pronounced as would be expected based on the ChP requirement for oleic acid only grade (≥98%)


Oxidative degradation studies (Fe(II)SO4)

➢ Chinese grade PS80 shows after 4 days exposure complete loss of polyester and monoester fractions

PS80 USP vs PS80 ChP after ≈4 daysat 40°C (Fe(II) sulfate)

14

UPLC-HRMS (pos mode) method to monitor oxidative degradation of 0.04% PS formulated at pH 6

➢ Chinese grade PS80 also with higher levels of oxidative degradants compared to multi-compendial PS80



➢ Chinese grade PS80 show a shorter lag-phase for the formation and higher levels

of oxidized degradants compared to multi-compendial grade

➢ PS20 shows lower levels compared to PS80 for intermittent oxidized ester PS

degradant

15

UPLC-HRMS (pos mode) method to monitor oxidative degradation of 0.04% PS formulated at pH 6 at 25°C

Oxidized ester PS degradant 1



➢ Some oxidized products are intermittent, others do not decompose during

monitored time period

➢ Free (“soluble”) oleic acid was not detected at high levels (either only present in

oxidized forms or as FA in mixed micelles or as emulsion at 25°C)

16

UPLC-HRMS (neg mode) method to monitor oxidative degradation of 0.04% PS formulated at pH 6 at 25°C

Koch et al. (2018), manuscript underpreparation“.

Oxidized FA PS degradant 2


Characterization of insoluble fraction of PS

Biopharmaceutical product stressed at 40°C

➢ Increase in subvisible particles after lag-time

➢ Smaller (≥10µm) particles increase first, then larger (≥25µm) particles

0.0

100.0

200.0

300.0

400.0

500.0

600.0

700.0

0 1 2 3 4 5 6 7

Nu

mb

er o

f p

arti

cles

/mL

Expsoure time in months

Subvisible particle analysis by light-obscuration

≥10µm particles/mL ≥25µm particles/mL

17

Examples of MFI

images

➢ Morphology of PS related particles is variable and not always easily

distinguishable from proteinaceous particles


Characterization of PS related (subvisible) particles

Isolation of particles on filter for advanced characterization of their

nature and composition by FTIR- microscopy and SEM-EDX

SEM image on filter surface

EDX Spectra: C, O major elements, trace metals(Au from filter surface)

18

FTIR Spectra: Fatty acid related nature


Quantitative methods to assess degradation of excipients (e.g., oxidation of histidine) or stabilizers (e.g., Met)ICP-MS for metal (drivers for oxidative stress)

Fast quantitative assay for major PS constituents with efficient removal of

protein (if possible automated)

Easy, fast, quantitative PS assay for surface active fraction of PS

Optimized LC-MS assay with sample preparation to

eliminate protein interferences and

advanced data analysis tools for efficient read-out

of complex PS mixture

Quantitative methods to assess number and size of particlesAdvanced characterization methods to determine nature and composition

Quantitative assay for free fatty acids in protein formulations to assess risk of particle formation

PS Analytical Toolbox with Complementary Assays

Reliable, automated assay to assess peroxides (incl. lipid species) in protein formulations to evaluate drivers and kinetics of oxidants in DP

PS content

PS class analysis

UPLC-HRMS PS characterization

assay

Fatty acid assayOxidant assay

Excipient assay

Particle characterization

tools

19


Janssen R&D, Pharmaceutical Development & Manufacturing Sciences

– René Spycher

– Jochen Büchler

– Mehul Patel

Coriolis Pharma

– Michelle Berger

– Constanze Helbig

– Ariadna Martos

Acknowledgments

20

complexity of the analytical characterization of polysorbate · 2018-04-02 · pharmaceutical...

Documents